Aalborg Universitet Whole Life Carbon Assessment of 60 buildings Possibilities to develop benchmark values for LCA of buildings Zimmermann, Regitze Kjær; Andersen, Camilla Marlene Ernst; Kanafani, Kai; Birgisdottir, Harpa

Aalborg Universitet

Whole Life Carbon Assessment of 60 buildings

Possibilities to develop benchmark values for LCA of buildings

Zimmermann, Regitze Kjær; Andersen, Camilla Marlene Ernst; Kanafani, Kai; Birgisdottir, Harpa

Creative Commons License Unspecified

Publication date:

2021

Document Version

Publisher's PDF, also known as Version of record Link to publication from Aalborg University

Citation for published version (APA):

Zimmermann, R. K., Andersen, C. M. E., Kanafani, K., & Birgisdottir, H. (2021). Whole Life Carbon Assessment of 60 buildings: Possibilities to develop benchmark values for LCA of buildings. Polyteknisk Boghandel og Forlag. BUILD Report No. 2021:12 https://sbi.dk/Pages/Whole-Life-Carbon-Assessment-of-60-buildings.aspx

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

- Users may download and print one copy of any publication from the public portal for the purpose of private study or research.

- You may not further distribute the material or use it for any profit-making activity or commercial gain - You may freely distribute the URL identifying the publication in the public portal -

Take down policy

BUILD REPORT 2021:12

WHOLE LIFE CARBON

ASSESSMENT OF 60 BUILDINGS

POSSIBILITIES TO DEVELOP BENCHMARK VALUES FOR LCA OF BUILDINGS

WHOLE LIFE CARBON

ASSESSMENT OF 60 BUILDINGS

POSSIBILITIES TO DEVELOP

BENCHMARKS VALUES FOR LCA OF BUILDINGS

Regitze Kjær Zimmermann, Camilla Ernst Andersen, Kai Kanafani & Harpa Birgisdóttir

BUILD

Department of the Built Environment Aalborg University Copenhagen

2021

TITLE Whole Life Carbon Assessment of 60 buildings

SUBTITLE Possibilities to develop benchmark values for LCA of build- ings

SERIES TITLE BUILD report

FORMAT PDF

YEAR OF PUBLICATION 2021 PUBLISHED DIGITALLY April 2021

AUTHORS Regitze Kjær Zimmermann, Camilla Ernst Andersen, Kai Ka- nafani & Harpa Birgisdóttir

LANGUAGE English

NUMBER OF PAGES 96

REFERENCES Page 61

KEYWORDS Sustainability, life cycle assessment, LCA, construction pro- cess stage, energy use, benchmark value, benchmark value, whole life carbon emission, Global Warming Potential, GWP ISBN

ISSN

978-87-93585-35-5 2597-3118 COVER ILLUSTRATION Michael Ulf Bech

PUBLISHED BY Polyteknisk Boghandel og Forlag ApS Anker Engelunds Vej 1

2800 Kongens Lyngby

This publication is subject to copyright law.

/The certification means that an independent peer of at least PhD level has made a written assessment justifying this book’s scientific quality and original contributions.

PREFACE

Sustainability is increasingly a key concept in the debate on quality assurance of buildings.

Sustainability within construction is about the environmental, economic and social quality of buildings, and is therefore considered a supplement to traditional qualities of a building.

There is extensive focus on finding solutions to reduce the carbon footprint of society.

This also applies to buildings and the construction industry as a whole. For a number of years, the construction industry has been using life cycle assessment (LCA) to document the environmental impact of buildings. In order to carry out an LCA of buildings, proper doc- umentation of the environmental impact of all materials used is necessary. In addition to documenting the environmental impact of materials, we are looking for different sustainable solutions to minimise the environmental impact and resource pressure of buildings.

The Danish Transport, Construction and Housing Authority has asked the Danish Build- ing Research Institute (now BUILD - The Department of the Built Environment, referred to as BUILD below) to carry out a number of projects as part of the increased focus on environ- mental sustainability, including the development of LCAbyg, a Danish LCA tool for buildings, launched in 2015. After a number of years of building competencies and compiling experi- ence within LCA, it is now possible to take the next step and examine how the environmen- tal impact of buildings can be reduced. This can be done by generating knowledge about the environmental impact of buildings and by developing benchmark values for LCA of buildings that can be used for legislation, common sector guidelines, DGNB certification or tender documents, for example.

The purpose of this report is to establish a more extensive knowledge base about the whole life carbon assessment of buildings that can be used to develop benchmarks values for buildings.

The report was prepared by BUILD in 2019 on behalf of the Danish Transport, Construction and Housing Authority. The report was prepared by Regitze Kjær Zimmermann, Camilla Ernst Andersen, Harpa Birgisdottir and Kai Kanafani. Before publication, the script was peer- reviewed by Morten Birkved, professor with special responsibilities at the University of Southern Denmark. BUILD is grateful for the constructive collaboration with Professor Morten Birkved.

BUILD – Department of the Built Environment (former Danish Building Research Institute), Aalborg University Copenhagen

Division of Energy Efficiency, Indoor Climate and Sustainability of Buildings March 2020

Søren Aggerholm Research Manager

SUMMARY

Globally, the building and construction sector is responsible for approx. 39% of all Green House Gas emissions, and approx. 28% comes from operational energy use for the total ex- isting building stock and approx. 11% comes from consumption of materials for new build- ings and refurbishment of existing buildings (World Green Building Council, 2019). Whole life carbon emissions (Global Warming Potential, GWP) and other environmental impacts from both operational energy use and from building materials can be determined and re- duced using life cycle assessments (LCA).

In Denmark, the Danish Transport, Construction and Housing Authority and The Depart- ment of the Built Environment (BUILD) have developed a tool to carry out LCA of buildings, LCAbyg, and have published a number of Danish publications about LCA of buildings. In practice, LCA has been used in the DGNB certification system for buildings since 2012.

Data from all these LCAs has not previously been collected and calculated according to a uniform method. Therefore, there is still a lack of broad understanding of the current level of environmental impacts from buildings. In addition, the DGNB does not yet include detached houses, accounting for a significant percentage of building activities in Denmark.

This report presents LCAs of 60 building cases built from 2013 to 2021. The case buildings come from DGNB-certified projects, external projects and life cycle assessments carried out by BUILD. The cases are divided into five building types focusing on homes and offices, see figure 1. When collecting the case buildings, attempts were made to include a broad selec- tion of cases with different qualities in terms of building types, energy classes, materials, photovoltaic area, etc. This takes into account the differences between buildings, so that the data basis for the benchmark values is as representative as possible.

FIGURE1. Number of cases by building type. There are 60 cases in total, 34 of which are homes and 26 are offices and other buildings – including a school, a hospital and multi-functional buildings.

The life cycle perspective from the LCA includes upfront carbon emissions, i.e. production of building materials, as well as impacts expected to take place on the basis of a future sce- nario related to replacements, operational energy and demolition. The calculations were made in LCAbyg with the associated calculation method and environmental database.

Results of the LCAs of all case buildings can be seen as bars in figure 2 for a 50-year refer- ence study period and an 80-year reference study period, respectively. The 50-year refer- ence study period is most widespread and in line with the European Level(s) reporting framework. The results show large variations in the impacts of buildings, as some buildings have up to 2.25 times greater impacts than others in a 50-year reference study period and up to 2.5 times greater impacts than others in an 80-year reference study period.

The impacts of the buildings can be divided into impacts from materials (referred to as embodied impacts) and impacts from operations. Impacts from materials are approx. 2-4 times greater than impacts from operations for a 50-year as well as an 80-year reference study period. Furthermore, there is a large gap between impacts from materials alone, vary- ing from 3.7 to 10.8 kg CO2 eq/m²/year at 50 years and 3.11 to 9.50 kg CO2 eq/m²/year at 80 years. The same applies to impacts from operational energy use, varying from 0.22 to 4.58 kg CO2 eq/m²/year at 50 years and 0.17 to 4.30 kg CO2 eq/m²/year at 80 years. Results also show that there is not a big difference in impacts for the GWP of different building types, neither in impacts for operational energy nor materials. The median value for the GWP of materials for detached houses, terraced houses, apartment buildings and offices is (7.4) (7.1) (7.0) and (6.9) kg CO2 eq/m²/year in a 50-year reference study period, respec- tively.

FIGURE 2. GWP and benchmark values of case buildings GWP is shown per square metre of gross floor area and year over a 50-year reference study pe- riod (top) and an 80-year reference study period (bottom).

The LCA results can be used to establish a common basis for the environmental perfor- mance of buildings by means of benchmark values. A common benchmark value can form the basis for tender requirements, public regulation or other types of benchmarking that al- ready exist for energy demand, indoor climate or other areas, but are missing for the life-cy- cle-based environmental impact in Denmark. Work on preparing an LCA can be facilitated by carrying out an estimated LCA. Annex III illustrates how the LCAbyg functions for an esti- mated LCA can be used to give a conservative estimate of the environmental impacts of buildings.

The results of the LCAs are included in a statistical analysis to determine the benchmark values on the basis of the 60 case buildings. The benchmark value is then expressed as the median, upper and lower quartile for a 50-year and an 80-year reference study period, re- spectively, each of which suggests a possible level of ambition. The median and quartiles are shown as horizontal lines in figure 2. The median value for the 50-year reference study period is 9.5 kg CO2 eq/m²/year, while the lower quartile is 8.5 kg CO2 eq/m²/year. However, the median value for the 80-year reference study period is 8.0 kg CO2 eq/m²/year, while the lower quartile is 6.9 kg CO2 eq/m²/year.

The figure also shows that several buildings range considerably below the lower quartile in both a 50-year and an 80-year reference study period. These buildings, with impacts be- low the lower quartile, can therefore also be included as benchmarks for buildings of the fu- ture. Results for the individual case buildings are in Annex IV.

0 2 4 6 8 10 12 14 16

A05 A06 A07 A08 E01 E02 E03 E04 E05 E06 E07 E08 E09 E10 E11 Enf01 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 K01 K02 K03 K04 K05 K06 K07 K08 K09 K10 K11 K12 K13 K14 K15 K16 K17 K18 K19 K20 K21 K22 R01 R02 R03 R04 R05 R06 R07 R08 R09 R10 R11 R12

kg CO2eq/m2/year

Case-bygninger median øvre kvartil nedre kvartil

0 2 4 6 8 10 12 14 16

A05 A06 A07 A08 E01 E02 E03 E04 E05 E06 E07 E08 E09 E10 E11 Enf01 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 K01 K02 K03 K04 K05 K06 K07 K08 K09 K10 K11 K12 K13 K14 K15 K16 K17 K18 K19 K20 K21 K22 R01 R02 R03 R04 R05 R06 R07 R08 R09 R10 R11 R12

kg CO2eq/m2/year

Case-bygninger median øvre kvartil nedre kvartil median

Case buildings upper

quartile lower quartile

Case buildings median upper

quartile lower quartile

The benchmark values in this report correspond to a bottom-up approach based on the performance of existing buildings. The 60 cases in this report represent the largest number of LCAs of buildings collected in Denmark to date. Furthermore, they have been compiled in the same calculation tool, LCAbyg, and are therefore based on the same environmental data and the same method of calculation. Variation in building type, materials, etc. also means that the data represents a broad selection of buildings in Denmark. This provides a sufficient basis for preparing benchmark values for voluntary schemes. The benchmark values should be updated as more LCAs of buildings become available.

CONTENTS

PREFACE 5

SUMMARY 6

1 INTRODUCTION 12

1.1 Background 12

1.2 Purpose 13

1.3 Reading guide 13

2 LCA OF BUILDINGS 15

3 CALCULATION BASIS 17

3.1 60 case buildings 17

3.2 Methodology for LCA 21

4 RESULTS FROM LCA IN A 50-YEAR REFERENCE STUDY PERIOD 26

4.1 Results from LCA of case buildings 26

5 RESULTS FROM LCA IN AN 80-YEAR REFERENCE STUDY PERIOD 33

5.1 Results from LCA of case buildings 33

6 PROMINENT CONDITIONS FOR WHOLE LIFE CARBON EMISSIONS AND

BENCHMARK VALUES 40

6.1 Reference study period 40

6.2 Building type and design 43

6.3 Photovoltaics 48

6.4 Energy class 50

6.5 Secondary buildings 53

6.6 Summary 54

7 POSSIBILITIES TO DEVELOP BENCHMARK VALUES 56

7.1 Possibilities to develop benchmark values 56

7.2 Benchmark values based on LCA of 60 case buildings 57

7.3 Benchmark-value updates 59

8 REFERENCES 61

ANNEX I: DESCRIPTION OF CASE BUILDINGS 63

ANNEX II: ADJUSTMENT FOR LACK OF DATA FOR THE TECHNICAL

INSTALLATIONS 67

ANNEX III: USE OF ESTIMATED LCA 74

ANNEX IV: DETAILED LCA RESULTS WITH 50-YEAR REFERENCE STUDY

PERIOD 76

ANNEX V: DETAILED LCA RESULTS WITH 80-YEAR REFERENCE STUDY

PERIOD 86

1 INTRODUCTION

1.1 Background

In recent years, interest in reducing anthropogenic impacts on the environment has in- creased, and the sustainable transition is now firmly on the agenda. Sustainability has so- cial, economic and environmental aspects, the latter being the focus of this report. Sustaina- bility is also considered an import aspect within the building and construction sector, and is increasingly an element in the quality assurance of buildings. As part of this development, the DGNB certification system was introduced in 2012 by the Green Building Council Den- mark. In 2014, a voluntary sustainability class in the Building Regulations was proposed in the building policy strategy (Danish Ministry of Climate, Energy and Building, 2014).

There is extensive focus on finding solutions to reduce the climate footprint of society.

The Danish government has set a goal to reduce GWP by 70% in the period from 1990 to 2030. This entails focus on identifying reduction possibilities in all corners of society, and the government has entered into a climate partnership with the business community in this re- spect.

The building and construction sector consumes a large percentage of the world's re- sources and contributes to negative impacts on the environment in the form of materials and energy use as well as generation of large volumes of waste. Globally, the building and con- struction sector is responsible for approx. 39% of all GWP, and approx. 28% comes from op- erational energy use for the total existing building stock and approx. 11% comes from con- sumption of materials for new buildings and refurbishment of existing buildings (World Green Building Council, 2019).

Generally, these environmental impacts can be brought down by reducing impacts from building materials and from the operation of buildings. For many years, Denmark has fo- cused on reducing impacts from operational energy use by regulating requirements for en- ergy demand in the Danish Building Regulations. This means that new buildings now have lower environmental impacts from operational energy use than from building materials, and therefore it is worth focusing on the environmental impact of building materials (Birgisdóttir &

Madsen, 2017).

Environmental impacts and resource use from both operation and building materials can be determined in an LCA of a building. Using an LCA, environmental impacts for a given building can be quantified and compared with similar buildings to assess how the environ- mental impacts can be reduced. In DGNB certification, LCA has been an important part of assessing environmental sustainability since 2012. The certification system is based on benchmark values for LCA that the building must comply with. Initially, German benchmark values were adapted and used as reference for the first Danish DGNB certifications. As Danish experience in carrying out LCA of buildings grew, Danish benchmark values were developed (Rasmussen, et al., 2019) (Rasmussen & Birgisdóttir, 2018). In Denmark, bench- mark values from DGNB have been used for LCA of buildings for about eight years, and they have been the largest source of Danish experience with LCA of buildings.

The proposed voluntary sustainability class in the building policy strategy from 2014 also included guidance efforts. Among other things, this meant that work was initiated to opera- tionalise the LCA of buildings. Consequently, the LCA tool, LCAbyg, was launched in 2015,

and since then various Danish publications on this area have been published, including In- troduction to LCA of Buildings (Birgisdottir & Rasmussen, 2015), LCA of Large Building Renovation (Birgisdottir & Rasmussen, 2015) Buildings' Embodied Energy and Environmen- tal Impacts (Birgisdóttir & Madsen, 2017) and Early Design Stage Building LCA (Kanafani, Zimmermann, Birgisdóttir, & Rasmussen, 2019). These publications and the growing num- ber of LCAs of Danish buildings have helped form the current experience and basis for com- parison, and this can be used to find the right level of reduction in the GWP of buildings.

France, Finland and Sweden are already developing national benchmark values for the carbon footprint of buildings. Moreover, several countries are considering introducing bench- mark values in their building regulations to regulate GWP. The Netherlands is the frontrun- ner, and has had declaration requirements since 2013 and statutory requirements on com- pliance with a stipulated limit value since 2018. In parallel, an international standard for the methodology to set benchmark values is being prepared (ISO 21678) and Annex 72 under the International Energy Agency (IEA) is working to establish a common understanding of the environmental impacts of buildings (Frischknecht R., Birgisdottir, Chae, Lützkendorf, T.,

& Passer, 2019). So, many initiatives have been launched regarding benchmark values for LCA of buildings, all of which are intended to illustrate the need to minimise environmental impacts in order to achieve a more sustainable building and construction sector.

1.2 Purpose

The purpose of this report is to establish an adequate data basis on the GWP of buildings in Denmark throughout their life cycle. On the basis of this data, possible benchmark values adapted to the LCA method used in Denmark are outlined. BUILD has previously prepared benchmark values for LCA for use in DGNB certification, but the benchmark values in this report are based on a considerably larger data basis and an updated method.

The data basis was developed to carry out LCA of 60 Danish case buildings. The report analyses and interprets the results and outlines a selection of benchmark values for the GWP. The selection of benchmark values can be used to set requirements to minimise the GWP of buildings, e.g. in legislation, DGNB certification or tender documents.

1.3 Reading guide

The report is divided into an introduction, method, results and analysis, and it culminates in the last section outlining benchmark values:

Chapter 2 of the report begins with a brief introduction to LCA of buildings.

Chapter 3 presents the case buildings and the LCA method used as the basis to calculate the benchmark values.

Chapter 4 presents the results from the LCAs on the selected case buildings over a 50-year reference study period. Possible benchmark values can be determined on the basis of these results.

Chapter 5 presents the results from the LCAs on the selected case buildings over an 80- year reference study period. Possible benchmark values can be determined on the basis of

Chapter 6 analyses the results in relation to parameters that have proven to be important in an LCA. There are analyses of how these parameters affect the results of an LCA and the benchmark values over a 50-year reference study period.

Chapter 7 outlines possible benchmark values for LCA of buildings for both a 50-year and an 80-year reference study period.

2 LCA OF BUILDINGS

LCA is a standardised method to assess potential environmental impacts and resource use of a building. The long-term perspective ensures that impacts from the full life cycle of the building are included, including the production of building materials, transport, installation, maintenance, replacements and processing of materials at the end-of-life stage and opera- tional energy for the building, see figure 3. In practice, Denmark does not yet include all stages of the life cycle (modules included are marked in blue in figure 3). This is due to fo- cus on the most environmentally important stages of the life cycle and a lack of experience and routines in documenting all other stages of the life cycle. This approach is also referred to as the simplified LCA in the European framework for sustainable buildings: Level(s). As experience grows and more LCAs of buildings become available, focus on including more life cycle stages is increasing – particularly the upfront stages such as the construction pro- cess stage.

FIGURE 3. Stages (A, B, C and D) and modules (A1, A2, etc.) in the life cycle of a building. The LCA quantifies environ- mental impacts for the individual stages or modules. The sum of environmental impacts constitutes the environmental profile of the building. The common practice in Denmark is only to include the modules marked in blue.

Environmental impacts and resource use are calculated separately for each stage of the life cycle, and on the basis of a reference study period. This period is assumed to constitute the operation stage of the building. At the end of the reference study period, the building will be considered demolished to complete the life cycle perspective. However, the reference study period should not be compared with the expected service life of the building, which may be longer.

Impacts are normally stated within a number of environmental impact categories reflect- ing the different types of environmental damage. Denmark usually operates with the follow- ing categories:

– Depletion Potential of the Stratospheric Ozone Layer (ODP) – Acidification Potential

– Eutrophication Potential

– Formation Potential of Tropospheric Ozone Photochemical Oxidants – Abiotic Depletion Potential for Non-fossil Resources

– Abiotic Depletion Potential for Fossil Resources – Total Use of Primary Energy

– Use of Renewable Secondary Fuels

If different environmental impact categories are to be used as one common benchmark value, this will require a decision on how these categories should be weighted against each other. This study focuses on GWP. GWP is an environmental impact indicator for the global warming potential of the earth's surface temperature on the basis of an increased concentra- tion of greenhouse gases contributing to the greenhouse effect. The unit is kg CO2 eq, where the various greenhouse gases are converted into the GWP of carbon dioxide.

The life cycle perspective includes upfront carbon emissions, i.e. production of building materials, as well as impacts expected to take place on the basis of a future scenario related to replacements, operational energy or demolition. This is illustrated in figure 4, in which the accumulated, i.e. combined, impacts are shown over a reference study period – here 50 years. The figure shows two curves: The upper curve relates to the impacts of materials, and the lower curve relates to impacts from operational energy.

Impacts from production of materials are seen in the large increase at year 0 of the up- per curve. Building parts are replaced in the period between year 0 and year 50. The re- placements result in impacts from disposal of the construction product and in impacts from production of a new, similar construction product. These impacts are seen as small and large increases on the upper curve. There are simultaneous impacts from operational en- ergy as illustrated by the lower curve. At year 50, impacts are calculated up to the end-of-life stage of the building, corresponding to demolition of the building and disposal of all building materials. These impacts can be seen as an increase in the upper curve.

The life cycle perspective is important to avoid staggering impacts from one stage of the life cycle to another. However, it is also important to be aware of upfront carbon emissions, particularly because these impacts can be calculated with greater certainty, and a reduction of these would have a direct environmental effect.

FIGURE 4. Accumulated impacts over the reference study period. The figure shows separate contributions from materi- als (upper curve) and operations (lower curve). The graph illustrates that buildings cause a significant impact from ma- terials during construction (A1-3). Over a reference study period, there will be impacts from replacement of materials (B4) and energy use (B6). In connection with demolition, there are impacts from processing the materials at the end-of- life stage (C3-4).

3 CALCULATION BASIS

3.1 60 case buildings

The data used to prepare benchmark values comes from DGNB-certified projects, external projects and life cycle assessments carried out by BUILD as part of this project. A total of 60 different case buildings have been included, and these have been/will be built between 2013 and 2021. They are divided into five building types (see figure 5). Figure 5 also shows a code for each building type to make it easier to identify the building types included in the re- sults in section 4, 0 and 6.

FIGURE 5. Number of cases by building type. There are 60 cases in total, 34 of which are homes and 26 are offices and other buildings – including a school, a hospital and multi-functional buildings.

Out of the 60 case buildings, 37 DGNB-certified projects have been included within the building types: Terraced houses, Apartment buildings and Offices. To ensure sufficient data and to include Detached houses, BUILD has carried out additional life cycle assessments of a number of projects. The assessments were based on material specifications from draw- ings obtained from architects, consultants and manufacturers of prefabricated houses. The remaining cases come from external projects obtained by BUILD. Among other things, the external projects concern buildings in the Other buildings category, and they include a school, a hospital and multi-functional buildings. When collecting the case buildings, at- tempts were made to include a broad selection of cases with different qualities in terms of building types, energy classes, materials, photovoltaic area, etc. This takes into account the differences between buildings, so that the data basis for the benchmark values is as repre- sentative as possible. However, no statistical assessment has been made of which case buildings are most representative of the Danish building stock.

As part of the data basis for the benchmark values, all projects are updated to LCAbyg version 4.0 (beta) in order to compensate for differences in the method and data base. This process ensures that all cases include the elements that LCAs should include according to

source of environmental data. LCAbyg version 4.0 (beta) is a beta version of the official LCAbyg version 3, and it is a calculation tool developed by BUILD and published by the Danish Energy Agency (now the Danish Transport, Construction and Housing Authority).

The DGNB-certified projects have been transferred from DGNB's LCA tool to LCAbyg, where the external projects and BUILD projects have been updated from an older version of LCAbyg to LCAbyg 4.0 (beta).

Figure 6 describes the building types according to source, energy class, area and con- struction type, and the construction type is stated in two categories – heavy and light build- ings. The differentiation between heavy and light buildings is related to the load-bearing structures, where heavy buildings have internal walls or concrete elements and light build- ings have skeleton constructions. The differentiation is independent of the type of façade cladding used.

FIGURE 6. Summary of the data basis for the five building types. Heavy buildings are defined as having load-bearing structures with internal walls or concrete elements, while light buildings have load-bearing structures in skeleton con- structions. See Annex I, table 8 for the distribution of the individual case buildings.

In order to create an overview of the primary materials in the cases, the materials for all case buildings have been examined and categorised within the following building parts:

• Foundation

• Basement slab

• Slabs

• External walls – load-bearing structures

• External walls – façades

• Internal walls,

• Windows

• Roof – load-bearing structures

• Roof – surface.

The Internal walls category includes both load-bearing and non-bearing internal walls, as it has not been possible to differentiate between these two categories. The materials have been further categorised into material groups covering the overall materials of the building.

These tend to vary from building to building and can significantly influence the environmental impacts. The categories for materials within the various building parts are:

• Materials containing concrete and cement

• Wood

• Metal

• Tiles (bricks and roof tiles)

• Roofing felt.

Table 1 shows the distribution of materials in the case buildings.

TABLE 1. Summary of materials in the building parts for all case buildings

3.2 Methodology for LCA

Life cycle stages

According to the assessment of environmental performance of buildings standard, EN 15978, an LCA has five different life cycle stages and 17 underlying modules (see figure 7).

Together, these make up the full life cycle of the building, taking into account consumption of building materials as well as processes regarding operation of a building (operational energy use and water use). In LCAbyg, it is currently only possible to calculate a selection of the 17 modules, i.e. production and transport of construction products (A1-3), replacement of build- ing parts (B4), operational energy (B6) and waste processing at the end-of-life stage (C3-4).

As LCAbyg version 4.0 (beta) has been chosen as the LCA tool to analyse the 60 case buildings, only the selected modules will be considered in this project. Figure 7 shows the modules included in EN 15978, as well as the modules covered by this project. See (Birgis- dottir & Rasmussen, 2015) for a general introduction to LCA of buildings.

FIGURE 7. Life cycle stages in EN 15978. The modules available in LCAbyg and considered in this project are marked with dark green.

Building parts included

An LCA aims for as complete a picture of the building as possible. DGNB-certified projects generally follow the rules from DGNB in terms of which building parts to include in an LCA (DK-GBC, 2016). For external projects, completeness highly depends on the data available at the time of modelling. Danish Building Research Institute projects generally follow the SfB classification system for building parts, in which the following building groups are included if present in the building:

- Foundations - Basement slabs - Slabs

- External walls

- Load-bearing structures - Internal walls

- Roofs

- Stairs/steps and ramps

- Balconies and access balconies - Windows, doors and glass façades - Drains

- Water - Heating

- Ventilation and cooling

- Electricity and mechanical facilities - Other

Technical installations

LCAs of buildings often lack data on technical installations. This also applies to the 60 case buildings in this report, where data for several groups of technical installations (drains, wa- ter, heating as well as ventilation and cooling) is missing for some of the case buildings. In some case buildings, these installations account for up to 10% of a building’s GWP of mate- rials (see Annex II, figure 27). As an incomplete building model could distort the result of the benchmark value, the results of the LCAs are adjusted for missing data on technical installa- tions. Technical installations are pipes and installations included in the overall groups:

drains, water, heating as well as ventilation and cooling. Photovoltaic modules are not in- cluded in technical installations, as there is sufficient data on photovoltaic modules. There- fore, it is not necessary to adjust the case buildings for missing data on photovoltaic mod- ules.

Adjustment was made by replacing all case building data on the technical installations mentioned with generic data on technical installations. The generic data is based on the cases containing sufficient data on technical installations. The case buildings with sufficient data generate generic values (the median value) for the GWP for the different groups of technical installations. The median value for the different groups of technical installations is shown in table 2.

TABLE 2. Calculation of the representative GWP (median) for technical installations.

Groups Projects with completed

group

Median value at 50 years [kg CO2 eq/m²/year]

Median value at 80 years [kg CO2 eq/m²/year]

Drains 18% 0.02 0.02

Water 23% 0.12 0.11

Heating 37% 0.23 0.24

Ventilation and cooling

30%

0.08

0.08

Sum 0.46 0.45

Note that the sum is calculated on the basis of the non-rounded values for medians. Therefore, the "Sum" row will not necessarily show the precise sum of the figures above.

Reference study period

A reference study period of 50 years and 80 years, respectively, is used for the LCAs of the 60 case buildings. Reference study period expresses the number of years the building is an- alysed in the LCA. The service life of the building can therefore be longer than the reference study period used. The longer the reference study period, the less the weight of the impacts during construction of the building. On the other hand, there will be greater weight on im- pacts in the use stage of the building, including replacement of materials and operational en- ergy use.

During the start of the DGNB system, Denmark decided to use the same reference study period for LCA for certification as in DGNB Germany and internationally; i.e. 50 years. Later, when the DGNB manuals were updated, it was decided to use short reference study periods of 50 years and longer reference study periods of 80 years for offices, 100 years for schools, institutions and hospitals, and 120 years for homes, in line with Danish Building Research Institute report no. 30 from 2013(Aagaard, Brandt, Aggerholm, & Haugbølle, 2013) The ref- erence study period of 50 years stems from financial depreciation periods of fixed asset in- vestments, whereas the longer periods reflect expectations of the actual service life of build- ings(Aagaard, Brandt, Aggerholm, & Haugbølle, 2013). An analysis of demolitions of 20,999

Danish buildings in the period 2009-2015 showed that the median value for the service life of the buildings was 59 years, and the average was 70 years (Østergaard, et al., 2018).

The general European and international practices normally use reference study periods of 50 to 60 years. In the preliminary version of Level(s), it was slightly unclear whether to use 50 or 60 years for LCA (Dodd, Cordella, Traverso, & Donatello, 2017). The most recent an- nouncement from Level(s) in February 2020 states that a reference study period of 50 years has been used in the updated version of Level(s). Table 3 shows reference study periods used in a comparative study of an office building carried out in the current international IEA Annex 72 project, which focuses on international harmonisation of LCA of buildings. Here, experts in 15 out of 21 countries used a reference study period of 50 years, 60 years in five out of 21 countries, and one country, i.e. Denmark, used a reference study period of 80 years. A comprehensive study of 650 scientific studies of LCAs of buildings shows that a ref- erence study period of 50 years is used in approx. 60% of all studies (Röck, et al., 2020).

Reference study periods of 80, 100 and 120 years are used in 9% of all studies.

The LCA results of the 60 case buildings have been examined for the significance of the reference study period in section 6.1.

TABLE 3.Reference study period used in an international comparative study of office buildings in the IEA Annex 72 project, which focuses on international harmonisation of LCA of buildings (Frischknecht R. , et al., 2019) (Frischknecht R., Birgisdottir, Chae, Lützkendorf, T., & Passer, 2019).

Belgium Brazil Canada Denmark France The Nether- Hong Kong Italy China New Zealand Norway Switzerland Portugal Spain United King- Sweden Czech Re- Germany Hungary USA Austria

50 x x x x x x x x x x x x x x x x

60 x x x x x

80 x

Replacement of construction products

The reference study period affects the replacement of construction products. Construction products with a shorter service life than the reference study period must be replaced one or several times during the reference study period. In this project, service life for the individual building parts will be based on SBi-2013:30 and service life is also available in LCAbyg. The number of replacements depends on the service life determined for the individual construc- tion products. In LCAbyg, it is assumed that construction products are only replaced if there are more than 10 years left of the reference study period and that construction products are not replaced if less than a third of the construction product's service life in the building is left.

Database

Life cycle assessments in this project are primarily based on materials available in LCAbyg version 4.0 (beta). The database in LCAbyg for materials mainly consists of generic or aver- age data from Ökobaudat 2016 and it only contains product-specific data to a limited extent.

Ökobaudat is a German database and is therefore not necessarily representative of Danish production in relation to environmental impacts and resource use. There is currently no Dan- ish database, and therefore it is not possible to use Danish data in the LCAs of buildings.

This means there is a risk that Danish data has a higher or lower environmental impact than that calculated, and therefore it would be best to use Danish data to estimate the exact envi- ronmental impacts.

Bio-based materials

The database in LCAbyg includes biogenic carbon in bio-based materials. Bio-based materi- als can capture, store and release carbon in their lifetime. This carbon is referred to as bio- genic carbon. The calculation of GWP for bio-based materials in the database in LCAbyg takes into account the capture and release of biogenic carbon, see EN 15804:2012. The standard states that GWP for bio-based materials should be calculated as negative in the product stage (modules A1-3) due to the capture of biogenic carbon during growth, and as positive at the end-of-life stage (modules C3-4) when the biogenic carbon is released in con- nection with decay or incineration. This means that the balance of biogenic carbon within the individual life cycle is calculated as 0. In this report, this means that case buildings contain- ing large quantities of bio-based materials will typically have a low or negative GWP in the product stage and a high positive GWP at the end-of-life stage. Note that according to the latest version of the product standard (EN 15804:2012+A2:2019) the stored biogenic carbon should be reported separately from the carbon related to fossil fuels and the carbon related to changes in land use. This division is not yet available in the data forming the basis for the calculations in this report.

Operational energy use

Impacts from operational use for all case buildings are calculated on the basis of data avail- able in LCAbyg. Data in LCAbyg is based on the report Nye emissionsfaktorer for el og fjern- varme (New emission factors for electricity and district heating) (COWI and the Danish Transport, Construction and Housing Authority, 2016). Projected data for the period 2015 to 2050 was chosen as a scenario in this project. This means that a gradual increase in the re- newable energy share in the energy grid is expected during the given period (2015-2050).

Environmental impact categories

Results of these analyses are shown in LCAbyg for nine different environmental indicators, all of which are standard indicators in EN 15978. However, the purpose of this project is to focus on the environmental indicator Global Warming Potential (GWP). This will therefore be the indicator in this report, and other indicators will be disregarded. The decision to focus on the GWP is based on the high priority of this topic today. An LCA will usually focus on sev- eral different environmental indicators in order to carry out a broad environmental assess- ment. It is essential to be aware of this, as other environmental indicators can be highly rele- vant and important if the full impact of a building is to be assessed.

Reference unit

Throughout the report, the results of LCAs will be presented in GWP, normalised to the area (per m²) and the reference study period (per year). Normalisation to per m² is when the im- pact from operational energy use is normalised over the heated gross floor area, and where impacts from the materials are normalised over the gross floor area. Among other things, this is to avoid diluting the impacts from operational energy over an area larger than the area to be heated. Furthermore, the reference study period is used to normalise the results to ‘per year’.

Data processing of results

The results have undergone simple statistical data processing, in which the main focus was to examine differences in the 60 case buildings, and how these differences could affect the results in GWP and thus any benchmark value. Section 6 includes different aspects which have proven to have great influence on the GWP examined, including photovoltaic modules.

Relevant aspects for which the influence is not so well known have also been examined, in- cluding building type and design, energy class and secondary buildings. The results of how these aspects influence the environmental impacts will depend on the case buildings used in the analysis. Therefore, they cannot be considered as conclusions applying to all buildings.

The aspects that have been examined more thoroughly are listed below.

- Reference study period - Building type and design - Photovoltaics

- Energy class - Secondary buildings

Furthermore, Annex III examines how estimated LCAs can influence GWP. Estimated LCAs are often used in the early design stage, when the design has not yet been fully defined.

4 RESULTS FROM LCA IN A 50-YEAR REFERENCE STUDY PERIOD

4.1 Results from LCA of case buildings

This section shows the results of all case buildings in a 50-year reference study period in kg CO2 eq/m² or kg CO2 eq/m²/year. The results are shown for all 60 case buildings, and ad- justments have been made for missing data on technical installations, as described in sec- tion 3.2.

Figure 8 shows the impacts from the case buildings calculated over a 50-year reference study period and shown per m²/year. The figure shows large variations in the total GWP of the buildings. Some buildings have up to 2.25 times greater impacts from both materials and operations than other buildings, varying from 6.45 to 14.52 kg CO2 eq/m²/year. Moreover, the figure shows that impacts from the building materials are typically 2-4 times higher than impacts from operational energy use. Impacts from materials vary from 3.67 to 10.84 kg CO2

eq/m²/year, whereas impacts from operational energy use vary from 0.22 to 4.58 kg CO2

eq/m²/year. In this context, it is important to note that the operational energy use for each building is based on data from energy performance framework calculations. The actual oper- ational energy use is usually higher, because the calculation method does not cover all con- sumption and uses standard assumptions. This means that the actual GWP is likely to be higher.

Similarly, the actual GWP of materials will also be higher. The section on methodology states that not all life cycle stages have been included in the calculation. This means that the calculation does not include impacts from transport to the construction site as well as instal- lation and material wasted on the construction site. Nor does it include repair and mainte- nance of building materials, which can also increase GWP.

Note that a single building (Enf11) has no data for operational energy use, and this build- ing therefore has no impact on operations (see figure 8).

FIGURE 8. GWP of the 60 case buildings over a 50-year reference study period broken down by embodied carbon emissions (materials) and operational carbon emissions. Enf11 has no data for operations, and therefore only results for materials are shown.

Figure 9 illustrates how impacts from materials are distributed on an annual basis. The re- sults are shown on a time axis in kg CO2 eq/m², and it is clear that some impacts are up- front, whereas others are part of a future scenario. The figure shows that the GWP of materi- als for most buildings is highest in year 0 when the building is constructed. However, some buildings have a low impact in year 0, but a high impact in year 50 when the reference study period ends. This is because these buildings have a greater share of wood products which

0 2 4 6 8 10 12 14 16

A05A06 A07A08 E01E02 E03E04 E05E06 E07E08 E09E10 Enf01E11 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11R01R02R03R04R05R06R07R08R09R10R11R12K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22

kg CO₂eq/m²/year Bygningsdele DriftOperation Embodied

carbon will be released again at the end-of-life stage, where the release is included in the LCA regardless of whether the wood is assumed to be incinerated, reused or recycled.

Buildings with a large share of biogenic material will therefore have a low or negative GWP in the product stage (modules A1-3) and a higher impact at the end-of-life stage (modules C3-4), which is also reflected in the results.

After upfront carbon emissions during construction of the building, replacement of materi- als (stage B4) is stated as impacts between year 0 and year 50. Figure 9 shows the that im- pacts from replacements happen in years 15, 20, 25 and 30, which usually corresponds to replacement of paint, roofing felt, double-glazed windows, photovoltaic modules and tech- nical installations. However, material consumption in year 0 (modules A1-A3) will still typi- cally result in the highest impact for most case buildings (see figure 10 to the left).

The figure also shows that, at the end of the reference study period in year 50, the total GWP of case buildings varies from 180 to 540 kg CO2 eq/m² when only considering impacts from materials. This shows that there is a potential to reduce total impacts per m² via the se- lection of materials.

Figure 10 to the right shows that impacts from materials primarily come from the building part groups roofs, external walls and slabs/basement slabs. In some case buildings, there is no differentiation between slabs, basement slabs and roof slabs, which means that for some cases there are no impacts from Roofs, as impacts from the roof slab is categorised under Slabs etc. This is due to different choices in the LCA and these differences are particularly evident in DGNB-certified and external projects.

Moreover, figure 10 shows large impacts from the groups Windows, Internal walls, Foun- dations and Photovoltaic modules (where these are included). This indicates that the large building part groups make up the largest share of total impacts for the case buildings, and that this is where the greatest potential exists to reduce environmental impacts from build- ings.

-300 -200 -100 0 100 200 300 400 500 600

0 5 10 15 20 25 30 35 40 45 50

kg CO2eq/m2

year Materials, 50 years

180 540

FIGURE 10. GWP of materials from the 60 case buildings over a 50-year reference study period broken down by life cycle stage (left) and building part group

-6 -4 -2 0 2 4 6 8 10

A05 A06 A07A08 E01 E02 E03E04 E05 E06E07 E08 E09E10 E11 Enf01 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22R01R02R03R04R05R06R07R08R09R10R11R12

kg CO₂eq/m²/year

A1-3B4 C3C4

0 2 4 6 8 10

A05 A06A07 A08 E01 E02E03 E04 E05E06 E07 E08E09 E10 E11 Enf01 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11R01R02R03R04R05R06R07R08R09R10R11R12K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22

kg CO₂eq/m²/year

Trapper og ramper Fundamenter søjler/bjælker Altaner og altangange Andet

Indervægge Dæk

El- og mekaniske anlæg (primært solceller) Vinduer, døre, glasfacader

Ydervægge Tage Terrændæk

Installationer for vand, varme, ventilation og afløb Stairs/steps and ramps

Foundations Beams/pillars

Balconies and access balconies Other

Internal walls Slabs

Electricity and mechanical facilities (primarily PV modules) Windows, doors and glass façades

External walls Roofs Basement slabs

Water, heating, ventilation and drainage installations

Figure 11 shows impacts from operational energy in kg CO2 eq/m². Again, the results are shown on a time axis as kg CO2 eq/m², showing that operational energy is decreasing over time. This is because the energy composition used in LCAbyg has been projected according to national goals for a gradually larger renewable energy share in the future, and this will have a lower GWP. This transition to renewable energy production has the greatest effect on electricity production. This means that buildings with high electricity consumption, espe- cially buildings with electric heat pumps, will have greater reductions in impacts over time compared with buildings with district heating as their heating source.

Figure 11 shows a significant spread in the results for operational energy use (from 11 to 230 kg CO2 eq/m² at 50 years) (see figure 11). However, the figure also shows that this spread includes two cases with a significantly lower operational energy use than the remain- ing case buildings, as well as a case with a significantly higher operational energy use. The case buildings with a low operational energy use only have contributions from electricity con- sumption, as these are heated with a heat pump. This results in low environmental impacts from operations over time. However, the high contribution from operational energy use is due to one building having a relatively high heating demand from district heating.

In general, the spread in GWP of operational energy use is partly due to the composition of energy, as described earlier, but also due to the size of the energy demand. Here, differ- ences in the energy frame and the possibility to obtain a supplement to the energy perfor- mance framework may influence the operational energy demand. Apart from the three ex- treme cases, impacts from operational energy use vary between 65 kg CO2 eq/m² and 154 kg CO2 eq/m² (over a 50-year reference study period) for the case buildings (see figure 11).

FIGURE 11. Accumulated GWP of operational energy use of the case buildings for a 50-year reference study period. GWP is stated per m² of heated gross

0 50 100 150 200 250

0 5 10 15 20 25 30 35 40 45 50

kg CO2 eq/m2

year Operational energy, 50 years

11 230

5 RESULTS FROM LCA IN AN 80-YEAR REFERENCE STUDY PERIOD

5.1 Results from LCA of case buildings

This section shows the results of all case buildings in an 80-year reference study period in kg CO2 eq/m² or kg CO2 eq/m²/year. The results include all 60 case buildings, and adjust- ments have been made for missing data on technical installations, as described in section 3.2.

Figure 12 shows the impact from the case buildings calculated over an 80-year reference study period and shown per m²/year. The figure shows large variations between total GWP of the case buildings, where some buildings have up to 2.5 times greater impacts from both materials and operations than other buildings (varying from 4.92 to 12.39 kg CO2

eq/m²/year). Moreover, the figure shows that impacts from the building materials are typi- cally 2-4 times greater than impacts from operational energy use. Impacts from materials vary from 3.11 to 9.50 kg CO2 eq/m²/year, whereas impacts from operational energy use vary from 0.17 to 4.30 kg CO2 eq/m²/year.

As is the case for the results of a 50-year reference study period, it is important to note that the actual impacts from operational energy use and from materials are likely to be higher than those calculated. Again, this is because the operational energy use is based on data from the energy performance framework calculation, and this is usually underestimated in relation to the actual consumption. Moreover, not all life cycle stages have been included in the calculation, and therefore the actual impact from materials will be higher.

It should be further noted that case building Enf11 has no impacts from operations, be- cause it has no data for operational energy use.

FIGURE 12. GWP of the 60 case buildings over an 80-year reference study period broken down by embodied carbon emissions (materials) and operational carbon emissions. Enf11 has no data for operations, and therefore only results for materials are shown.

Figure 13 illustrates how impacts from materials are distributed on an annual basis. The re- sults are shown on a time axis in kg CO2 eq/m², and it is clear that some impacts are up- front, whereas others are part of a future scenario. The figure shows that the GWP of materi- als for most buildings is highest in year 0 when the building is constructed. In line with the 50-year reference study period, some buildings have a low or negative impact in year 0, but

0 2 4 6 8 10 12 14

A05A06 A07A08 E01E02 E03E04 E05E06 E07E08 E09E10 Enf01E11 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11R01R02R03R04R05R06R07R08R09R10R11R12K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22

kg CO₂eq/m²/year Bygningsdele Drift Embodied Operation

a high impact in year 80 when the reference study period ends. This is because these build- ings have a greater share of wood products which store biogenic carbon, resulting in a low impact in the product stage (see section 4.1 and 3.2 for a more detailed explanation).

Replacement of materials (stage B4) is stated as impacts between year 0 and year 80.

Figure 13 shows that the impacts from replacements happen in years 15, 20, 25, 30, 40, 50 and 60, which usually corresponds to replacement of paint, roofing felt, double-glazed win- dows, photovoltaic modules, technical installations and surfaces (façade material, floors and ceilings).

The figure also shows that, at the end of the reference study period in year 80, the total GWP of materials for all case buildings varies from 250 to 760 kg CO2 eq/m². This shows that there is a potential to reduce total impacts per m² via the selection of materials.

Figure 14 to the right shows that impacts from materials primarily come from the building part groups roofs, external walls and slabs/basement slabs. As is the case for the 50-year reference study period, some case buildings do not differentiate between slabs, basement slabs and roof slabs. This is because of different choices in the LCA, which is typical in DGNB-certified and external projects (see section 4.1).

Moreover, figure 14 shows large impacts from the groups Windows, Internal walls, Foun- dations and Photovoltaic modules (where these are included). Once again this emphasises that, regardless of reference study period, the large building part groups account for the highest share of total impacts for the case buildings, and that this is where the greatest po- tential exists to reduce environmental impacts from buildings.

FIGURE 13. Accumulated GWP of materials of case buildings over an 80-year reference study period. GWP is stated per m² of gross floor area. GWP for operations is not included in the graph.

-400 -200 0 200 400 600 800 1000

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

kg CO2 eq/m2

year Materials, 80 years

250 760

-4 -2 0 2 4 6 8

A05 A06 A07 A08E01 E02 E03E04 E05 E06E07 E08 E09 E10E11 Enf01 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22R01R02R03R04R05R06R07R08R09R10R11R12

kg CO₂eq/m²/year

A1-3 B4 C3 C4

0 2 4 6 8 10

A05A06 A07 A08E01 E02 E03 E04E05 E06 E07 E08E09 E10 Enf01E11 Enf02 Enf03 Enf04 Enf05 Enf06 Enf07 Enf08 Enf09 Enf10 Enf11R01R02R03R04R05R06R07R08R09R10R11R12K01K02K03K04K05K06K07K08K09K10K11K12K13K14K15K16K17K18K19K20K21K22

kg CO₂eq/m²/year

Trapper og ramper Fundamenter søjler/bjælker Altaner og altangange Andet

Indervægge Dæk

El- og mekaniske anlæg (primært solceller) Vinduer, døre, glasfacader

Ydervægge Tage Terrændæk

Installationer for vand, varme, ventilation og afløb Stairs/steps and ramps

Foundations Beams/pillars

Balconies and access balconies Other

Internal walls Slabs

Electricity and mechanical facilities (primarily PV modules) Windows, doors and glass façades

External walls Roofs Basement slabs

Water, heating, ventilation and drainage installations